У нас вы можете посмотреть бесплатно 4. Spurious Correlation или скачать в максимальном доступном качестве, видео которое было загружено на ютуб. Для загрузки выберите вариант из формы ниже:
Если кнопки скачивания не
загрузились
НАЖМИТЕ ЗДЕСЬ или обновите страницу
Если возникают проблемы со скачиванием видео, пожалуйста напишите в поддержку по адресу внизу
страницы.
Спасибо за использование сервиса ClipSaver.ru
A correlation is a correspondence between events, such as the tendency for birthday cakes to be served on someone’s birthday. The word “spurious” basically means “fake.” So a spurious correlation is a “fake” correlation. In what sense is a spurious correlation fake? We often infer a causal relation between correlated events. For instance, one might assume that cake is being served because it’s my birthday. But that assumption may be false—if, for instance, the cake is being served for some other reason, and people don’t even know it’s my birthday. To guard against spurious correlations, students of science are taught that “correlation is not causation.” This does not mean that correlation never implies causation. Rather, it means that correlations sometimes point to a causal relation, but sometimes not. Google “spurious correlations,” and you’ll find lots of amusing examples of correlated events that are not causally related. For instance, ice cream sales and shark attacks are correlated: when ice cream sales rise, shark attacks do too. But a moment’s reflection shows that my choice to buy ice cream is not causing anyone to be attacked by sharks. So, why does the correlation exist? Both ice cream sales and shark attacks rise when the weather heats up. Warmer weather causes us to desire ice cream, and also to go swimming in the ocean—where the sharks live. Thus, both phenomena are caused by another phenomenon (changing weather), and there is not a direct causal relation between ice cream sales and shark attacks. Although it is easy to see that the correlation between sharks and ice cream is spurious, many spurious correlations are more insidious. For instance, until I went to work in a cognitive neuroscience lab and designed my own experiments using fMRI, I had no idea that fMRI brain imaging data are purely correlational—this means, brain activity that is correlated with a task might indicate a causal involvement of that brain region in the task—but it might not: Activity in the region might just be a spurious correlation. It’s a little easier to appreciate this point if we envision the brain as encompassing the entire neural system, including nerves throughout the body. I may consistently snap my fingers while I sing a tune, but we shouldn’t infer that the neural activity in my fingers is causing me to sing. Similarly, just because a region of the brain lights up during a task doesn’t necessarily mean that this region is performing the task. The fallacy of seeing spurious correlations can be thought of as part of a broader tendency of human minds to impose meaningful connections on the world, even when these connections are not inherently meaningful. Have you ever noticed that after you buy a new car, you start seeing the same model of car all over town? This happens because your attention has become attuned to see the car you’ve just purchased. You aren’t really encountering more of that model—you’re just noticing it more. I have a good friend who started seeing 11:04 when she was young. After she told me and the rest of our friend, we all started seeing 11:04. Of course, we weren’t really seeing the number more than before—but we were noticing it more. The event of seeing 11:04 was just a coincidence, but because we were noticing this event more often, our human minds imposed meaning onto it. In order to counteract this natural tendency to over-interpret our world, imposing artificial meaningful onto reality, it is important to appreciate that a coincidence is not necessarily evidence, and correlation does not necessarily imply causation.